CN108513089A - The method and device of group's video session - Google Patents
The method and device of group's video session Download PDFInfo
- Publication number
- CN108513089A CN108513089A CN201710104442.4A CN201710104442A CN108513089A CN 108513089 A CN108513089 A CN 108513089A CN 201710104442 A CN201710104442 A CN 201710104442A CN 108513089 A CN108513089 A CN 108513089A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- head
- model
- virtual portrait
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/14—Systems for two-way working
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/14—Systems for two-way working
- H04N7/141—Systems for two-way working between two video terminals, e.g. videophone
- H04N7/147—Communication arrangements, e.g. identifying the communication as a video-communication, intermediate storage of the signals
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of method and devices of group's video session, belong to VR (Virtual Reality, virtual reality) technical field.This method includes：The virtual portrait of the first user in group's video session is obtained, the virtual portrait of the first user is obtained according at least to the head feature data of the first user and the corresponding limbs model of the first user；During group's video session, the behavioural characteristic data of virtual portrait and the first user based on the first user obtain the video data of the first user, in video data the action of the virtual portrait of the first user matched with the actual act of the first user；The video data that the first user is sent to terminal where the second user for participating in group's video session, to realize group's video session.The present invention can express the reality image of the first user more agilely, enhance visual effect when group's video session.
Description
Technical field
The present invention relates to VR (Virtual Reality, virtual reality) technical field, more particularly to a kind of group's video councils
The method and device of words.
Background technology
VR technologies are a kind of technologies that can be created with the experiencing virtual world, the environment true to nature that can be simulated and intelligently
Perceive the behavior of user so that user feels on the spot in person.Therefore, application of the VR technologies in terms of social activity receives extensive pass
Note, the method for carrying out group's video session based on VR technologies are come into being.
In group's video session, multiple virtual portraits can be provided in VR equipment so that use the virtual use of VR equipment
Family can voluntarily select a virtual portrait, and in turn, Virtual User can express the image of oneself by the virtual portrait, to
It is exchanged with other Virtual User in group video session.
In the implementation of the present invention, the inventor finds that the existing technology has at least the following problems：
The reality image of each Virtual User has respective feature, and the virtual portrait provided in VR equipment is limited, very
May be far from each other with the authentic image of Virtual User, cause the effect of virtual portrait expression Virtual User poor, group's video council
Visual effect when words is poor.
Invention content
In order to solve problems in the prior art, an embodiment of the present invention provides a kind of methods and dress of group's video session
It sets.The technical solution is as follows：
On the one hand, a kind of method of group's video session is provided, the method includes：
The virtual portrait of the first user in group's video session is obtained, the virtual portrait of first user is according at least to institute
The head feature data and the corresponding limbs model of first user for stating the first user obtain；
During group's video session, virtual portrait and first user based on first user
Behavioural characteristic data obtain the video data of first user, the virtual portrait of the first user described in the video data
Action matched with the actual act of first user；
The video data that first user is sent to terminal where the second user for participating in group's video session, with
Realize group's video session.
On the other hand, a kind of device of group's video session is provided, described device includes：
Virtual portrait acquisition module, the virtual portrait for obtaining the first user in group's video session, described first uses
The virtual portrait at family is obtained according at least to the head feature data and the corresponding limbs model of first user of first user
It arrives；
Video data acquisition module is used for during group's video session, the void based on first user
The behavioural characteristic data of anthropomorphic object and first user, obtain the video data of first user, in the video data
The action of the virtual portrait of first user is matched with the actual act of first user；
Sending module, for sending first user to terminal where the second user for participating in group's video session
Video data, to realize group's video session.
The embodiment of the present invention is by the virtual portrait of the first user in group's video session, and the virtual portrait is according to
The head feature data of one user and corresponding limbs model obtain so that the virtual portrait can match the reality with the first user
Border image, moreover, having obtained the video data of first user based on the virtual portrait and behavioural characteristic data so that first uses
The action of the virtual portrait at family can also match the actual act of the first user, to express the reality of the first user more agilely
Image enhances visual effect when group's video session.
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings
Attached drawing.
Fig. 1 is a kind of implementation environment schematic diagram of group's video session provided in an embodiment of the present invention；
Fig. 2 is a kind of method flow diagram of group's video session provided in an embodiment of the present invention；
Fig. 3 is a kind of flow chart obtaining virtual portrait provided in an embodiment of the present invention；
Fig. 4 is a kind of flow chart obtaining head positioning data provided in an embodiment of the present invention；
Fig. 5 is a kind of flow chart obtaining video data provided in an embodiment of the present invention；
Fig. 6 is a kind of flow chart of group's video session provided in an embodiment of the present invention；
Fig. 7 is a kind of flow chart of display video data provided in an embodiment of the present invention；
Fig. 8 is a kind of device block diagram of group's video session provided in an embodiment of the present invention；
Fig. 9 is a kind of block diagram of the device 900 of group's video session provided in an embodiment of the present invention.
Specific implementation mode
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
Fig. 1 is a kind of implementation environment schematic diagram of group's video session provided in an embodiment of the present invention.Referring to Fig. 1, the reality
Applying environment includes：
At least one conventional terminal 101 (e.g., mobile terminal and desktop computer), at least one VR equipment 102 and at least one
A server 103.Wherein, server 103 be used to obtain the user for using conventional terminal 101 or VR equipment 102 virtual portrait,
Virtual portrait and behavioural characteristic data based on the user obtain video data, video data is sent to conventional terminal 101 or
VR equipment 102 so that group's video session can be carried out between different types of user.Conventional terminal 101 is for receiving and showing
Show the video data that server 103 is sent.VR equipment 102 can be used for obtaining virtual portrait, the base of the user of VR equipment 102
Video data is obtained in the virtual portrait and behavioural characteristic data of the user, sends video data to another VR equipment 102, or
Person can be also used for receiving the video data transmitted by simultaneously display server 103.
In addition, the server 103 can also configure at least one database, such as face model database, limbs pattern number
According to, virtual portrait database, user profile database and customer relationship chain database etc..The face model database is for depositing
The face model in card storage Tonghua；Limbs model database is used to store the limbs model of cartooning, and the limbs model database is also
Dressing can be stored with；User identifier and virtual portrait of the virtual portrait database for corresponding storage user；Subscriber data number
At least it is used to store the user properties such as age data, gender data and the occupation data of user according to library；Customer relationship chain database
The customer relationship chain data having for storing user, e.g., it is good friend that customer relationship chain data, which are at least used to indicate with the user,
The user of relationship or group relation.
It should be noted that when VR equipment 102 from server 103 for that when obtaining virtual portrait, can configure at least
Face model, limbs model or virtual portrait are obtained in one database.Moreover, the virtual portrait involved in the embodiment of the present invention
(including head model and limbs model) is three dimensional form.
Fig. 2 is a kind of method flow diagram of group's video session provided in an embodiment of the present invention.Referring to Fig. 2, this method can
To be applied to server or VR equipment, using server as executive agent for, this method specifically includes：
201, the virtual portrait of the first user in group's video session is obtained.
Group's video session refers to the video session that multiple (two or more) users are carried out based on server.Its
In, multiple users can be multiple users in the corresponding social platform of the server, may be group between multiple user
Relationship or friend relation.It should be noted that the user in group's video session can be the Virtual User using VR equipment,
Can also be the legacy user using conventional terminal (e.g., desktop computer, mobile phone).
First user can be any user in group's video session.The virtual portrait of first user is according at least to
The head feature data of one user and the corresponding limbs model of the first user obtain.The embodiment of the present invention is to obtaining virtual portrait
Opportunity does not limit.For example, when server is that multiple users create group's video session, obtains and each used in multiple user
The virtual portrait at family.In another example during group's video session, which receives a certain in group's video session
The invitation of user so that when server determines that group's video session is added in first user, obtain the visual human of first user
Object.
In the embodiment of the present invention, the head feature data and corresponding limbs model that server can be according to the first user,
It is that the first user creates virtual portrait in real time, to get the virtual portrait.Alternatively, the virtual portrait number of server configuration
According to the virtual portrait that may also have prestored the first user in library, therefore server can also be marked according to the user of the first user
Know, inquiry is with the presence or absence of virtual portrait corresponding with user identifier in virtual portrait database, if it is, can directly obtain
The virtual portrait of first user is got, if it is not, then can be that first user creates virtual portrait in real time.In fact,
Pre-stored virtual portrait is also to be created by server in virtual portrait database, be that is to say, the mistake of virtual portrait is obtained
Journey includes establishment process.Wherein, following steps 201A-201D may be used in the process that virtual portrait is got based on establishment process
It is obtained：
201A, the head feature data for obtaining the first user.
The head feature data are used to describe the practical head image of first user, can serve to indicate that first user
Hair zones, natural hair color, face area, face's tone, in face position and face form at least one of.Wherein, five
Official's form includes at least face tone and face profile.
The embodiment of the present invention does not limit the mode for obtaining head feature data.Such as：
Server obtains the head image data of the first user, analyzes the tone of head image data, obtains
To head characteristic.The source of the head image data can there are many, e.g., the head image in the cloud photograph album of the first user
The head image data of data (major part photograph) or the camera current shooting of the first user.Certainly, server can also obtain
The head image of multiple the first users, to more fully analyze head image data.In fact, server can also provide bat
Prompt is taken the photograph, which prompts for prompting user to be shot with different shooting angle so that server can be got not
With the head image data of shooting angle, so that the reality image of the head model and the first user that subsequently obtain is more
Match.
Since the tone of the hair of user, face and face respectively has feature, (e.g., yellow's color development is generally black, face
Generally partially yellow, eye is black and white, and mouth is red), and tone light and shade (e.g., the part relatively outstanding such as mouth, the bridge of the nose and forehead
General brighter, the wing of nose and eye socket are generally dark) also have nothing in common with each other, therefore, server can obtain head spy based on features described above
Levy data：
When determining face's tone and face area, server can be by the color value of the pixel in head image data
Be compared with configured a variety of colours of skin, if it exceeds the color value of the continuous image vegetarian refreshments of the first ratio with a certain colour of skin
The colour of skin, then can be determined as face's tone, and the image-region that matched continuous image vegetarian refreshments is constituted is determined as by matching
Face area.
When determining natural hair color and hair zones, server can determine the continuous image vegetarian refreshments adjacent with face area
For hair zones, and the color value of the continuous image vegetarian refreshments is extracted as natural hair color.
When determining face position, since the tone of mouth, eyes and eyebrow is different from face's tone, server can incite somebody to action
Hollow area in determining face area is identified as mouth, eyes and eyebrow position.Wherein, the position of eyebrow is located at most
Above, followed by eyes, mouth are located at bottom.Moreover, because ear protrudes outward with respect to face, server can determine
The edge pixel point of the both sides of face area, analyzes the tangent slope of the edge pixel point, if from pixel A to pixel B
The change rate of tangent slope be satisfied by default change rate, then the region where pixel A to pixel B can be determined as ear
Portion position.In addition, since nose is more three-dimensional with respect to face, generally nose both sides and it is following can form shade, and the bridge of the nose is bright
Degree is higher, therefore server can analyze shading value in face area and be higher than the continuous image vegetarian refreshments of the first shading value, and is located at
The continuous image vegetarian refreshments of the continuous image vegetarian refreshments both sides, the continuous image vegetarian refreshments of lower section shading value be less than the second shading value, by this three
Region where dividing continuous image vegetarian refreshments is determined as nose shape.According to the face position of above-mentioned determination, server can be according to five
The shape that edge pixel point where official position is constituted is determined as face profile, by the color of the pixel where face position
It is determined as face tone, to obtain face form.Certainly, in order to characterize the three-dimensional degree of nose, server can record nose
Higher than the shading value ratio of the pixel of the first shading value and the pixel less than the second shading value, the shading value ratio in sub- position
Example is higher, shows that the nose of the first user is more three-dimensional.
In fact, the mode for obtaining head feature data above is merely exemplary, the embodiment of the present invention can also use
Any mode obtains head feature data, for example, the recognizer based on face template or being identified using neural network
Algorithm.
Certainly, server can also continue to be modified determining head feature data, for example, according to the head feature
Face position in data determines face ratio, and the face ratio and configured normal face ratio are compared, if
Normal face ratio is not met, being adapted to property of server corrects the position of certain part in face so that face ratio meets
Normal face ratio.In fact, the normal face ratio is used to indicate the range residing for normal face ratio, therefore comparing
In the process, which meets the range residing for normal face ratio.
It should be noted that in order to save the calculation resources of server, server can also analyze necessary head feature number
According to necessary head feature data are used to be briefly described the practical head image of first user, e.g., necessary head feature data
It can serve to indicate that face's tone, face position and face form.
201B, according to head feature data, generate the head model with head feature Data Matching.
Based on the head feature data that step 201A is got, in order to more meticulously express the head model of virtual portrait,
Its head model is set more to be matched with the practical appearance of the first user, which can be specially：According to face area and hair
Region determines that contouring head model, contouring head model include face mask model and hair skeleton pattern；According to face's tone
And natural hair color, fill face mask model and hair skeleton pattern；It obtains and the matched face model of face form；According to five
Face model is blended into face mask model, generates the head model with head feature Data Matching by official position.
For example, server determines that face's tone is milky, natural hair color is brown, then server can be according to facial regions
The shape that the edge pixel point in domain (hair zones) is constituted is determined as face mask (hair profile), to generate face mask mould
Type (hair skeleton pattern), so that it is determined that contouring head model, in turn, server can fill face mask mould with milky
Type obtains facial model, fills hair skeleton pattern with brown, obtains Hair model.Further, server can be by nose
The face forms such as form, mouth form are compared with the face model to cartoonize in face model database, acquisition and face
The highest face model of tone, face contour similarity, and according to face position, the face model of acquisition is respectively synthesized to
On the face mask model of filling, according to the radian of face mask model and hair skeleton pattern, three-dimensional head model is built,
So that the head model generated is matched with the practical head image of the first user.
In fact, the face model that server can also cartoonize according to face form generating, for example, in mouth form
Mouth tone fill mouth profile, and deepen the pixel on the both ends line of mouth profile, generate mouth model, and mouth
Model is in " two valves " effect.For example, the eye tone in eye form includes at least two kinds, i.e. eyeball tone and eye white tone,
Eye white tone is generally inclined white tone, and therefore, server is the inclined white tone filling eye profile that can be used in eye tone, is used
The ball-type profile in another tone filling eye profile in eye tone, the ball-type profile and eye profile are tangent.
It should be noted that in order to more meticulously express the head image of the first user, server can also be further
Handle the head model.For example, server, which is Hair model, adds texture, and the age data of first user is obtained, in face
The texture of addition and the age-matched of first user on portion's model.In another example server obtains the gender number of first user
According to if first user is women, the eyelash length on eye model can be extended, reinforce the brightness of mouth model.Again
For example, server obtains the occupation data of the first user, if first user is student, can be added in facial model
Glasses model.
201C, according to the user property of the first user, determine the corresponding limbs model of the first user.
Wherein, user property is not limited to gender, age and the occupation of user.Usually, user can note in social platform
User property is filled in when volume account so that server can obtain user property, and deposit user property is corresponding with user identifier
Storage.
Since the reality image of user is often closely related with the user properties such as gender, age, occupation, height, weight, because
This, to make virtual portrait be more in line with the reality image of the first user, server can be obtained according to the user identifier of the first user
The corresponding user property of the user identifier is taken, in turn, according to user property, selection and user property from limbs model database
Matched limbs model.Moreover, server can also provide dressing model.
Wherein, the embodiment of the present invention does not limit the mode for providing dressing.For example, may include in the limbs model
Dress, alternatively, dressing model can also be provided separately in server, which can be stored in limbs model database, also may be used
To be stored in the dressing model database of server configuration.If dressing model is provided separately in server, can incite somebody to action die-filling
Type and corresponding dressing option are supplied to the first user so that the first user can be selected corresponding die-filling by dressing option
Type.Alternatively, server can also obtain the image data of the first user, the clothes that the first user of this in image data is worn are determined
Dress, matches dressing model corresponding with the clothes that the first user is worn, which is supplied to the first user.It is matching
When dressing model, it is not limited to be matched according to clothing color or shape.It is somebody's turn to do alternatively, server can be determined according to user property
The dressing model of first user, detailed process are similar with following determining processes of limbs model.
In addition, if limbs model includes dressing model, server can also use following at least three kinds of user properties
Determine limbs model：
(1), according to the gender data of the first user, the matched limbs model of gender data with the first user is determined.
Usually, male's stature is more strong, and women stature is more small and weak, therefore, can be with needle in limbs model database
The characteristics of to male and female stature, provides a variety of sexs dedicated limbs model, and each limbs model corresponds to one
Gender label so that server can determine the matched limbs of gender data with first user according to gender label
Model, moreover, the dressing of the limbs model of male's label can be fitted pants, the dressing of the limbs model of women label can be skirt
Dress.
(2), according to the age data of the first user, the matched limbs model of age data with the first user is determined.
Usually, if age of user is bigger, the garment language of the user can be more ripe.Therefore, limbs model data
It can be directed to the age bracket belonging to user in library, the limbs model of a variety of garment languages is provided, each limbs model corresponds to one
Age bracket label, for example, it is under-18s to have the corresponding age bracket label of the limbs model of case of caricatures of persons in dressing so that service
Device can determine the limbs model that the age data with first user meets according to age bracket label.
(3), according to the occupation data of the first user, the matched limbs model of occupation data with the first user is determined.
In real life, the business wear of the user of different occupation is also different, therefore, in limbs model database
A variety of limbs models for wearing business wear can also be provided, each limbs model corresponds to a professional label, for example, Western-style clothes limbs
The corresponding professional label of model is white collar, and the corresponding professional label of school uniform limbs model is student so that server can basis
Professional label determines the limbs model that the occupation data with first user meets.
It should be noted that the embodiment of the present invention does not limit the form of the corresponding label of each limbs model.For example,
Each limbs model can correspond to above-mentioned at least two label simultaneously, alternatively, the corresponding label of each limbs model is simultaneously
With two layers of meaning, e.g., which is schoolmarm's label.Once limbs model corresponds at least two labels or corresponding label tool
There is two layers or more of meaning, can allow server according at least two user properties, determine the corresponding limb of the first user
Body Model.For example, gender data and occupation data of the server according to the first user, determine that first user is doctress, then
It is women and the limbs model that professional label is doctor, or lookup mark that gender label can be searched from limbs model database
Label are the limbs model of doctress, and the limbs model found can be determined as to the corresponding limbs model of first user.
It should be noted that when determining limbs model, in addition to according to user property, group's video session can also be referred to
Corresponding group type, the virtual environment in group's video session and current actual temperature.The group type refers to the group
The group type of multiple affiliated groups of user in group video session.Explanation respectively is determined into limbs with reference to above-mentioned three kinds of data below
The concrete mode of model：
Method of determination 1, server determine the group type of multiple affiliated groups of user in group's video session, will be with group
The limbs model of type matching is determined as the limbs model with the first user.For example, each limbs model corresponds to group's class
Type label, Western-style clothes limbs model can correspond to company's cluster label, therefore, when the group type is group of company, server
The corresponding Western-style clothes limbs model of company's cluster label can be found, Western-style clothes limbs model is determined as to the limbs mould of the first user
Type.
Method of determination 2, server determine the corresponding virtual environment type of group's video session, will be with virtual environment type
The limbs model matched is determined as the limbs model of the first user.For example, the type of the virtual environment is sandy beach, then server can be with
Beach suit limbs model is determined as the corresponding limbs model of first user.
Method of determination 3, server obtain current actual temperature, will be true with the current matched limbs model of actual temperature
It is set to the limbs model of the first user.For example, current actual temperature is 35 degree, then server can be true by summer clothing limbs model
It is set to the corresponding limbs model of first user.
In fact, when server determines the limbs model of the first user, or the first user provides adjustment option.This
Inventive embodiments are not specifically limited the mode of adjustment option and offer adjustment option.For example, server determines the first user
Initial limbs model after, by initial limbs model and adjustment option be supplied to the first user, which includes height tune
Whole option, shape adjustment option and dressing adjust option, and the first user can adjust option adjustment height by triggering height
Just, fat or thin, the triggering dressing adjustment option of triggering shape adjustment option adjustment build replace dressing.
It should be noted that step 201C is the optional step of the embodiment of the present invention, in fact, due to head model foot
To characterize the reality image of first user, in order to realize that process is simple, the calculation resources of server are reduced, it can also be according to
The gender data of one user selects one and the matched limbs model of gender data from limbs model database at random.
In addition, it is necessary to which explanation, the embodiment of the present invention do not limit the sequential of above-mentioned steps 201A and 201C.It is true
On, server can also first determine limbs model, alternatively, server determines head model and limbs model simultaneously.
201D, head model and limbs model are synthesized, obtains the virtual portrait of the first user.
By step 201, server obtains the head image data of user, has carried out face and hair technical finesse,
Face and facial feature localization are obtained, generates head model according to face model data and limbs model database etc., and determine limb
Body Model synthesizes head model on the top of limbs model, to obtain a complete virtual portrait.Referring to Fig. 3,
An embodiment of the present invention provides a kind of flow charts obtaining virtual portrait.
It should be noted that the visual effect of the virtual portrait to make is more preferable, server can also be tied in synthesis
Close the ratio of head model and limbs model.For example, height data and configured normal person of the server according to the first user
Head body ratio data, determine the synthesis size of head model and limbs model, and by head model and limbs model adjust to
Determining synthesis size, then carry out the process of synthesis virtual portrait so that obtained virtual portrait is more in line with the first user
Reality image.In fact, to keep virtual portrait more attractive, server can also synthesize the virtual portrait of " Q editions ", " Q
The virtual portrait of version " is the virtual portrait for the head body ratio that finger body ratio does not meet normal person.Usually, to make " Q editions "
Virtual portrait is more lovely, and a body ratio data can be exaggerated more, and e.g., head body ratio data is 1:1.Server can be by
According to the head body ratio data of configured " Q editions ", determine the synthesis size of head model and limbs model, and by head model and
Limbs model is adjusted to determining synthesis size, then is synthesized, to obtain the virtual portrait of " Q editions ".
202, during group's video session, the behavioural characteristic of virtual portrait and the first user based on the first user
Data obtain the video data of the first user.
Wherein, behavioural characteristic data are used to indicate the actual act of first user, include at least expressive features data, mouth
It is any in type characteristic, head orientation characteristic and expression in the eyes direction character data.Pass through above step 201, server
Static virtual portrait is got, in the embodiment of the present invention, to make the virtual portrait mobilism, server obtain the first user's
Video data, and the action of the virtual portrait of the first user is matched with the actual act of the first user in the video data.This hair
Bright embodiment does not limit the mode for obtaining the video data.For example, being based on above-mentioned at least four behavioural characteristics data, this hair
Bright embodiment provides following at least four modes for obtaining video data：
When acquisition modes 1, behavioural characteristic data include expressive features data, when server detects the expression of the first user
When characteristic is specified expressive features data, limbs characteristic corresponding with specified expressive features data is obtained；It will specify
Expressive features data map to the head model of the virtual portrait of the first user in real time, and limbs characteristic is mapped in real time
The limbs model of the virtual portrait of first user, obtains the video data of the first user.
To make virtual portrait more meet the current reality image of the first user, the form of the first user of figurative expression, clothes
Being engaged in device can be by specified expressive features data and limbs characteristic Joint Mapping to virtual portrait.In the acquisition modes, service
Device can obtain the image data that the camera of the first user takes in real time, mark and in tracking image data face area and
The pixel or face area of face position and the crucial pixel of face position, to capture the expression of first user
Characteristic, crucial pixel is for describing face position and face form basicly.In turn, server can compare the table
The pixel distribution of feelings characteristic, the pixel distribution with specified expressive features data, the specified expressive features data refer to
The configured expressive features data of server, each specified expressive features data, which correspond to, configures a limbs characteristic, if
The similarity of the two reaches predetermined threshold value, then detects that the expressive features data are specified expressive features data.
By taking specified expressive features data open characteristic greatly for mouth as an example, if in the image data that server captures
The pixel distribution that characteristic is opened greatly in the pixel distribution of mouth position with mouth matches, and can obtain and the big Zhang Tezheng of mouth
The corresponding hand of data seals mouth characteristic, and therefore, server can be mouth model foundation three-dimensional coordinate, on three-dimensional coordinate
The pixel distribution for opening the pixel distribution adjustment mouth model of characteristic instruction greatly according to mouth, thus by the big Zhang Te of mouth
Sign data map to the mouth model in head model；Similarly, server can also seal the instruction of mouth characteristic according to hand
The pixel distribution of pixel distribution adjustment arm models, to which hand is sealed the hand that mouth characteristic maps in limbs model
Arm model so that virtual portrait mobilism, and then obtain the video data of the first user.
Using specified expressive features data as sobbing expressive features data instance, if in the image data that server captures
The pixel distribution of eye locations is matched with the distribution of the pixel of sobbing expressive features data, and can also obtain and cry expression spy
The corresponding hand of sign data rubs one's eyes characteristic, and sobbing expressive features data are mapped to the eye model in head model, and
Rub one's eyes the pixel distributions of the pixel distribution adjustment arm models of characteristic instruction according to hand, to which hand rub one's eyes spy
Sign data map to the arm models in limbs model.
It should be noted that make the reasonable transition of the image in video data, server can also be in continuous multiple frames video
Gradually mouth model and the corresponding pixel distribution of arm models are adjusted in data, to obtain to reflect that virtual portrait acts
The multi-frame video data of variation.
The acquisition modes in the figural expressive features data of reality for detecting user and configured by specifying
When expressive features Data Matching, the limbs characteristic with specified expressive features Data Matching is obtained, and is the virtual of the user
Personage, which assigns, specifies expressive features and limbs feature, to obtain video data, due to not allowing when user itself wears VR equipment
Own self emotion is directly easily expressed by limb action, which not only enables the practical table of virtual portrait analog subscriber
Feelings can more predict the mood of the user by expressive features, and the mood of expression user is protruded with limbs feature, to simultaneously
The figure image of analog subscriber in such a way that expression and limb action are united so that the expressive force and authenticity of virtual portrait are more
By force.
When acquisition modes 2, behavioural characteristic data include nozzle type characteristic, server is by the nozzle type characteristic of the first user
The head model that the virtual portrait of the first user is mapped to when factually obtains the video data of the first user.
Mouth action when to make the first user of video data synchronization of the first user make a speech, when server receives first
When the audio data of user, configured nozzle type characteristic is obtained, which is used to indicate mouth and is continuously in
The nozzle type characteristic can be mapped to the mouth model in head model in real time by folding condition, in turn, server, and by sound
Frequency with the virtual portrait after mapping according to being synthesized, to obtain the video data of the first user, until receiving audio data
Process terminate, server cancels the process of mapping mouth model, and mouth model is restored to default conditions, the default conditions
It refer to the state that mouth model remains closed.
When acquisition modes 3, behavioural characteristic data include head orientation characteristic, server obtains the sensing of the first user
The head positioning data of collected first user of device；The head orientation characteristic of first user is mapped into the first use in real time
The head model of the virtual portrait at family obtains the video data of the first user.
In order to make virtual portrait more vivo express the reality image of the first user, server can obtain first in real time
Sensor (e.g., nine axle sensors in VR equipment) collected head positioning data of user, the head positioning data is at least
It is used to indicate the pitch angle or left-right rotary corner of the first user, in turn, server can be according to indicated by head positioning data
Pitch angle or left-right rotary corner, the limbs model relative to the virtual portrait rotate the head model, thus by head orientation feature
The data map to head model in real time.
Certainly, to keep the head positioning data of acquisition more accurate, server can be combined with the camera of the first user
The image data taken, with reference to Fig. 4, an embodiment of the present invention provides a kind of flow charts obtaining head positioning data.Service
Device can obtain the image data that camera captures, and changed according to the pixel of face area in image data, work as facial regions
When the pixel in domain intensively offsets to one side, determine that head is in deflection state, and the negative direction of offset direction is determined as
Head deflection direction (the case where for self-timer), and deflection angle is determined according to the offset of pixel, to obtain head side
Position characteristic.When combining above two to obtain the mode of head orientation characteristic, server can determine two heads
Data error between orientative feature data can re-start acquisition head orientation if data error is more than fault-tolerant error
The process of characteristic, if data error is less than fault-tolerant error, the mode that data fusion may be used obtains head feature number
According to e.g., taking the average value of head feature data as correct head feature data.
Acquisition modes 4, behavioural characteristic data include expression in the eyes direction character data, and server obtains the camera of the first user
The eyes image data of the first user taken；According to the eyes image data of the first user, the expression in the eyes of the first user is obtained
Direction character data；The expression in the eyes direction character data of first user are mapped to the head mould of the virtual portrait of the first user in real time
Type obtains the video data of the first user.
For the interaction between each user in enhancing group video session, server can also obtain expression in the eyes direction character number
According to the expression in the eyes direction character data are used to indicate position of the eyeball with respect to eye of the first user, and then can serve to indicate that
The expression in the eyes gaze-direction of one user.
Since eyeball is different with the tone of the white of the eye, server can lock the eyeball in eyes image data, and real
When tracking eyeball with respect to eye position, to get expression in the eyes direction character data.Further, server can root
According to the expression in the eyes direction character data, the eyeball position in eye model is adjusted, and generate and obtain video data, thus by expression in the eyes side
The eye model in head model is mapped to characteristic.
The acquisition modes obtain the expression in the eyes direction character data of user by the eyes image data taken, thus will
The expression in the eyes direction character data of user map to the head model of the virtual portrait of the first user in real time.Not only so that virtual portrait
More meticulously real person's image of performance user so that virtual portrait is more matched with real person's image of user, and
And it can enhance expression in the eyes of each user in group's video session on the basis of showing each user expression in the eyes details and hand over
Stream improves the efficiency of group's video session.
In fact, the obtained video data of step 202 can be used as the initial video data of the first user, in order to give group
Second user in video session provides and the matched video data in its visual angle, and server can also be further to initial video number
According to being handled, for example, referring to Fig. 5, an embodiment of the present invention provides a kind of flow chart obtaining video data, server obtains
The perspective data of second user；Visual angle indicated by perspective data according to second user, handles initial video data,
Obtain the video data with matched first user in the visual angle.
Wherein, the embodiment of the present invention does not limit the mode for obtaining perspective data.For example, server can be according to second
The collected head orientation characteristic of sensor of user, obtains the corresponding visual angle of head orientation characteristic of second user
Data.In the citing, server is according to head positioning data, it may be determined that postrotational head model is oriented second user
Visual angle, to get the perspective data of second user.
In another example the eyes image data that server takes according to the camera of second user, obtain second user
Expression in the eyes direction character data obtain the perspective data of second user according to the expression in the eyes direction character data of second user.The citing
In, server can be directed toward eyeball position according to the eyeball position indicated by expression in the eyes direction character data with the center of head model
The direction set is determined as the visual angle of second user, to get the perspective data.
In turn, server can determine the visual angle indicated by the perspective data first based on the perspective data of second user
Field range in beginning video data, to extract the video counts of the video data within the vision as the first user
According to.With reference to Fig. 6, an embodiment of the present invention provides a kind of flow chart of group's video session, in group's video session, server
It can be by obtaining virtual portrait, and the face and face of the first user of real-time tracking, to get real-time video data,
And the video data is sent to terminal where second user in real time.
203, the video data that the first user is sent to terminal where the second user for participating in group's video session, to realize
Group's video session.
In the embodiment of the present invention, for any user in group's video session, server can according to step 201 and
202 obtain the video data of the user, and therefore, for the virtual portrait of each user of simultaneous display, server can be with composite group
The video data of each user in group video session, the video data after synthesis is sent to terminal where second user.When
, can be with real-time display video data when terminal where two users receives video data, and the video data and second user
Visual angle matches, to realize group's video session.With reference to Fig. 7, an embodiment of the present invention provides a kind of streams of display video data
Cheng Tu, server handle initial video data by obtaining initial video data, according to the perspective data of second user, will handle
Obtained video data is sent to terminal where second user so that terminal where second user being capable of regarding according to second user
Angle real-time display video data.It should be noted that execution master of the VR equipment as the embodiment of the present invention where the first user
When body, which can be sent to server, where being sent the video data to second user by server eventually
End.
The embodiment of the present invention is by the virtual portrait of the first user in group's video session, and the virtual portrait is according to
The head feature data of one user and corresponding limbs model obtain so that the virtual portrait can match the reality with the first user
Border image, moreover, having obtained the video data of first user based on the virtual portrait and behavioural characteristic data so that first uses
The action of the virtual portrait at family can simulate the actual act of the first user in real time, to express the first user's more agilely
Reality image enhances visual effect when group's video session.
Additionally, it is provided obtaining the concrete mode of virtual portrait, according to head feature data, generate and head feature data
Matched head model, and according to the user property of the first user, determine limbs model corresponding with the first user, pass through synthesis
Head model and limbs model obtain virtual portrait, have refined the acquisition process of virtual portrait each section so that virtual portrait has
There is more careful feature, to more meticulously express the reality image of the first user.Moreover, the limbs model is according to user
Attribute obtains, the reality image for making virtual portrait more be close to the users.
Additionally, it is provided the concrete mode of head feature data is obtained, the color of the head image by analyzing the first user
Distribution is adjusted, determines the head feature data of the first user, and the head feature data may be used to indicate the hair area of the first user
Domain, natural hair color, face area, face's tone, face position and face form, to obtain the practical head of the first user
The multinomial feature of image, can more careful, comprehensively describe the practical head image of the first user.
Additionally, it is provided generate the detailed process with the head model of head feature Data Matching, according to face area and
Hair zones determine face mask model and hair skeleton pattern, are filled according to face's tone and natural hair color, and according to
Face position will be blended into face mask model with the matched face model of face form, refine the mistake for generating head model
Journey, and the generating process of each part matches with the practical head image of the first user in head model, to improve
The matching degree of virtual portrait and first user's reality image.
Additionally, it is provided the mode of the limbs model of at least three kinds the first users of determination, according to the gender of the first user, year
The user properties such as age or occupation determine the matched limbs model of user property with the first user, moreover, these three methods of determination
It can also be combined, limbs model is not only made to be more in line with the reality image of the first user, but also make the side for determining limbs model
Formula is more diversified.
In addition, specifically illustrating when behavioural characteristic data include expressive features data, the video counts of the first user are obtained
According to concrete mode can obtain when it is specified expressive features data to detect expressive features data and specify expression special with this
The corresponding limbs characteristic of data is levied to map limbs characteristic to which specified expressive features data are mapped to face
To limbs model so that the expression-form of the virtual portrait of the first user is more lively.
In addition, specifically illustrating when behavioural characteristic data include nozzle type characteristic, head orientation characteristic and eye
When refreshing direction character data, the concrete mode of the video data of the first user is obtained, not only enables virtual portrait more vivo
The reality image of the first user is expressed, and makes the mode for obtaining the first video data more diversified.
Additionally, it is provided the visual angle indicated by perspective data according to second user, handles the mode of initial video data,
To obtain the video data with matched first user in the visual angle of second user so that show the first user's for second user
The visual angle of virtual portrait more meets actual visual effect.
Additionally, it is provided at least two obtain the mode of the perspective data of second user, according to the sensor of second user
Collected head orientation characteristic, or the eyes image data that are taken according to the camera of second user, depending on
Angular data can not only obtain the visual angle of second user in real time, and make the mode diversification for obtaining perspective data.
Fig. 8 is a kind of device block diagram of group's video session provided in an embodiment of the present invention.Referring to Fig. 8, the device is specific
Including：
Virtual portrait acquisition module 801, the virtual portrait for obtaining the first user in group's video session, the first user
Virtual portrait obtained according at least to the head feature data of the first user and the corresponding limbs model of the first user；
Video data acquisition module 802 is used for during group's video session, the virtual portrait based on the first user
With the behavioural characteristic data of the first user, the video data of the first user is obtained, the virtual portrait of the first user in video data
Action matched with the actual act of the first user；
Sending module 803, for sending regarding for the first user to terminal where the second user for participating in group's video session
Frequency evidence, to realize group's video session.
The embodiment of the present invention is by the virtual portrait of the first user in group's video session, and the virtual portrait is according to
The head feature data of one user and corresponding limbs model obtain so that the virtual portrait can match the reality with the first user
Border image, moreover, having obtained the video data of first user based on the virtual portrait and behavioural characteristic data so that first uses
The action of the virtual portrait at family can simulate the actual act of the first user in real time, to express the first user's more agilely
Reality image enhances visual effect when group's video session.
Optionally, virtual portrait acquisition module 801 is used for：Obtain the head feature data of the first user；According to head spy
Data are levied, the head model with head feature Data Matching is generated；According to the user property of the first user, the first user couple is determined
The limbs model answered；Head model and limbs model are synthesized, the virtual portrait of the first user is obtained.
Optionally, virtual portrait acquisition module 801 is used for：Obtain the head image data of the first user；To head image
The tone of data is analyzed, and head feature data are obtained, and head feature data are used to indicate the hair area of the first user
Domain, natural hair color, face area, face's tone, face position and face form.
Optionally, virtual portrait acquisition module 801 is used for：According to face area and hair zones, contouring head mould is determined
Type, contouring head model include face mask model and hair skeleton pattern；According to face's tone and natural hair color, face is filled
Skeleton pattern and hair skeleton pattern；It obtains and the matched face model of face form；According to face position, face model is closed
At to face mask model, the head model with head feature Data Matching is generated.
Optionally, virtual portrait acquisition module 801 is used for：According to the gender data of the first user, determine and the first user
The matched limbs model of gender data；And/or virtual portrait acquisition module 801 is used for：According to the age number of the first user
According to the determining matched limbs model of age data with the first user；And/or virtual portrait acquisition module 801 is used for：According to
The occupation data of first user determines the matched limbs model of occupation data with the first user.
Optionally, behavioural characteristic data include expressive features data, and video data acquisition module 802 is used for：When detecting
When the expressive features data of first user are specified expressive features data, it is special to obtain limbs corresponding with specified expressive features data
Levy data；Specified expressive features data map to the head model of the virtual portrait of the first user in real time, and by limbs feature
Data map to the limbs model of the virtual portrait of the first user in real time, obtain the video data of the first user.
Optionally, behavioural characteristic data include nozzle type characteristic, and video data acquisition module 802 is used for：First is used
The nozzle type characteristic at family maps to the head model of the virtual portrait of the first user in real time, obtains the video counts of the first user
According to.
Optionally, behavioural characteristic data include head orientation characteristic, and video data acquisition module 802 is used for：It obtains
The head positioning data of collected first user of sensor of first user；Factually by the head orientation characteristic of the first user
When map to the first user virtual portrait head model, obtain the video data of the first user.
Optionally, behavioural characteristic data include expression in the eyes direction character data, and video data acquisition module 802 is used for：It obtains
The eyes image data for the first user that the camera of first user takes；According to the eyes image data of the first user, obtain
Take the expression in the eyes direction character data of the first user；The expression in the eyes direction character data of first user are mapped to the first user's in real time
The head model of virtual portrait obtains the video data of the first user.
Optionally, video data acquisition module 802 is used for：The behavior of virtual portrait and the first user based on the first user
Characteristic obtains the initial video data of the first user；Obtain the perspective data of second user；According to the visual angle of second user
Visual angle indicated by data, handles initial video data, obtains the video data with matched first user in visual angle.
Optionally, video data acquisition module 802 is used for：It is special according to the collected head orientation of the sensor of second user
Data are levied, the corresponding perspective data of head orientation characteristic of second user is obtained；Or, video data acquisition module 802 is used
In：The eyes image data taken according to the camera of second user obtain the expression in the eyes direction character data of second user, root
The perspective data of second user is obtained according to the expression in the eyes direction character data of second user.
The alternative embodiment that any combination forms the present invention may be used, herein no longer in above-mentioned all optional technical solutions
It repeats one by one.
It should be noted that：Above-described embodiment provide group's video session device in group's video session, only with
The division progress of above-mentioned each function module, can be as needed and by above-mentioned function distribution by not for example, in practical application
Same function module is completed, i.e., the internal structure of device is divided into different function modules, to complete whole described above
Or partial function.In addition, the device for group's video session that above-described embodiment provides and the method for group's video session are implemented
Example belongs to same design, and specific implementation process refers to embodiment of the method, and which is not described herein again.
Fig. 9 is a kind of block diagram of the device 900 of group's video session provided in an embodiment of the present invention.For example, device 900 can
To be provided as a server.With reference to Fig. 9, device 900 includes processing component 922, further comprises one or more processing
Device, and by the memory resource representated by memory 932, for store can by the instruction of the execution of processing component 922, such as
Application program.The application program stored in memory 932 may include it is one or more each refer to corresponding to one group
The module of order.In addition, processing component 922 is configured as executing instruction, to execute the side of group's video session in Fig. 2 embodiments
Method.
Device 900 can also include the power management that a power supply module 926 is configured as executive device 900, and one has
Line or radio network interface 950 are configured as device 900 being connected to network and input and output (I/O) interface 958.Dress
Setting 900 can operate based on the operating system for being stored in memory 932, such as Windows ServerTM, Mac OS XTM,
UnixTM,LinuxTM, FreeBSDTMOr it is similar.
One of ordinary skill in the art will appreciate that realizing that all or part of step of above-described embodiment can pass through hardware
It completes, relevant hardware can also be instructed to complete by program, the program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (22)
1. a kind of method of group's video session, which is characterized in that the method includes：
Obtain the virtual portrait of the first user in group video session, the virtual portrait of first user is according at least to described the
The head feature data of one user and the corresponding limbs model of first user obtain；
During group's video session, the behavior of virtual portrait and first user based on first user
Characteristic, obtains the video data of first user, and the virtual portrait of the first user described in the video data moves
Work is matched with the actual act of first user；
The video data that first user is sent to terminal where the second user for participating in group's video session, to realize
Group's video session.
2. according to the method described in claim 1, it is characterized in that, described obtain the virtual of the first user in group video session
Personage includes：
Obtain the head feature data of first user；
According to the head feature data, the head model with the head feature Data Matching is generated；
According to the user property of first user, the corresponding limbs model of first user is determined；
The head model and the limbs model are synthesized, the virtual portrait of first user is obtained.
3. according to the method described in claim 2, it is characterized in that, the head feature data packet for obtaining first user
It includes：
Obtain the head image data of first user；
The tone of the head image data is analyzed, the head feature data, the head feature number are obtained
According to hair zones, natural hair color, face area, face's tone, face position and the face shape for being used to indicate first user
State.
4. according to the method described in claim 2, it is characterized in that, the head of the generation and the head feature Data Matching
Model includes：
According to the face area and the hair zones, contouring head model is determined, the contouring head model includes face
Skeleton pattern and hair skeleton pattern；
According to face's tone and the natural hair color, the face mask model and the hair skeleton pattern are filled；
It obtains and the matched face model of the face form；
According to the face position, the face model is blended into the face mask model, is generated and the head feature
The head model of Data Matching.
5. according to the method described in claim 2, it is characterized in that, the user property according to first user, determines
The corresponding limbs model of first user includes：
According to the gender data of first user, the matched limbs model of gender data with first user is determined；With/
Or,
According to the age data of first user, the matched limbs model of age data with first user is determined；With/
Or,
According to the occupation data of first user, the matched limbs model of occupation data with first user is determined.
6. according to the method described in claim 1, it is characterized in that, the behavioural characteristic data include expressive features data, institute
The behavioural characteristic data of virtual portrait and first user based on first user are stated, regarding for first user is obtained
Frequency is according to including：
When it is specified expressive features data to detect the expressive features data of first user, obtain and the specified expression
The corresponding limbs characteristic of characteristic；
The specified expressive features data are mapped to the head model of the virtual portrait of first user in real time, and will be described
Limbs characteristic maps to the limbs model of the virtual portrait of first user in real time, obtains the video of first user
Data.
7. according to the method described in claim 1, it is characterized in that, the behavioural characteristic data include nozzle type characteristic, institute
The behavioural characteristic data of virtual portrait and first user based on first user are stated, regarding for first user is obtained
Frequency is according to including：
The head model that the nozzle type characteristic of first user is mapped to the virtual portrait of first user in real time, obtains
To the video data of first user.
8. according to the method described in claim 1, it is characterized in that, the behavioural characteristic data include head orientation characteristic
According to the behavioural characteristic data of the virtual portrait and first user based on first user obtain described first and use
The video data at family includes：
Obtain the head positioning data of collected first user of sensor of first user；
The head orientation characteristic of first user is mapped to the head mould of the virtual portrait of first user in real time
Type obtains the video data of first user.
9. according to the method described in claim 1, it is characterized in that, the behavioural characteristic data include expression in the eyes direction character number
According to the behavioural characteristic data of the virtual portrait and first user based on first user obtain described first and use
The video data at family includes：
Obtain the eyes image data for first user that the camera of first user takes；
According to the eyes image data of first user, the expression in the eyes direction character data of first user are obtained；
The expression in the eyes direction character data of first user are mapped to the head mould of the virtual portrait of first user in real time
Type obtains the video data of first user.
10. according to the method described in claim 1, it is characterized in that, the virtual portrait and institute based on first user
The behavioural characteristic data of the first user are stated, the video data for obtaining first user includes：
The behavioural characteristic data of virtual portrait and first user based on first user obtain first user's
Initial video data；
Obtain the perspective data of the second user；
The visual angle indicated by perspective data according to the second user, handles the initial video data, obtain with
The video data of matched first user in visual angle.
11. according to the method described in claim 10, it is characterized in that, the perspective data packet for obtaining the second user
It includes：
According to the collected head orientation characteristic of the sensor of the second user, the head side of the second user is obtained
The corresponding perspective data of position characteristic；Or,
The eyes image data taken according to the camera of the second user, the expression in the eyes direction for obtaining the second user are special
Data are levied, the perspective data of the second user is obtained according to the expression in the eyes direction character data of the second user.
12. a kind of device of group's video session, which is characterized in that described device includes：
Virtual portrait acquisition module, the virtual portrait for obtaining the first user in group's video session, first user's
Virtual portrait is obtained according at least to the head feature data and the corresponding limbs model of first user of first user；
Video data acquisition module is used for during group's video session, the visual human based on first user
The behavioural characteristic data of object and first user obtain the video data of first user, described in the video data
The action of the virtual portrait of first user is matched with the actual act of first user；
Sending module, for sending regarding for first user to terminal where the second user for participating in group's video session
Frequency evidence, to realize group's video session.
13. device according to claim 12, which is characterized in that the virtual portrait acquisition module is used for：
Obtain the head feature data of first user；
According to the head feature data, the head model with the head feature Data Matching is generated；
According to the user property of first user, the corresponding limbs model of first user is determined；
The head model and the limbs model are synthesized, the virtual portrait of first user is obtained.
14. device according to claim 13, which is characterized in that the virtual portrait acquisition module is used for：
Obtain the head image data of first user；
The tone of the head image data is analyzed, the head feature data, the head feature number are obtained
According to hair zones, natural hair color, face area, face's tone, face position and the face shape for being used to indicate first user
State.
15. device according to claim 13, which is characterized in that the virtual portrait acquisition module is used for：
According to the face area and the hair zones, contouring head model is determined, the contouring head model includes face
Skeleton pattern and hair skeleton pattern；
According to face's tone and the natural hair color, the face mask model and the hair skeleton pattern are filled；
It obtains and the matched face model of the face form；
According to the face position, the face model is blended into the face mask model, is generated and the head feature
The head model of Data Matching.
16. device according to claim 13, which is characterized in that
The virtual portrait acquisition module is used for：According to the gender data of first user, determine with first user's
The matched limbs model of gender data；And/or
The virtual portrait acquisition module is used for：According to the age data of first user, determine with first user's
The matched limbs model of age data；And/or
The virtual portrait acquisition module is used for：According to the occupation data of first user, determine with first user's
The matched limbs model of occupation data.
17. device according to claim 12, which is characterized in that the behavioural characteristic data include expressive features data,
The video data acquisition module is used for：
When it is specified expressive features data to detect the expressive features data of first user, obtain and the specified expression
The corresponding limbs characteristic of characteristic；
The specified expressive features data are mapped to the head model of the virtual portrait of first user in real time, and will be described
Limbs characteristic maps to the limbs model of the virtual portrait of first user in real time, obtains the video of first user
Data.
18. device according to claim 12, which is characterized in that the behavioural characteristic data include nozzle type characteristic,
The video data acquisition module is used for：
The head model that the nozzle type characteristic of first user is mapped to the virtual portrait of first user in real time, obtains
To the video data of first user.
19. device according to claim 12, which is characterized in that the behavioural characteristic data include head orientation characteristic
According to the video data acquisition module is used for：
Obtain the head positioning data of collected first user of sensor of first user；
The head orientation characteristic of first user is mapped to the head mould of the virtual portrait of first user in real time
Type obtains the video data of first user.
20. device according to claim 12, which is characterized in that the behavioural characteristic data include expression in the eyes direction character number
According to the video data acquisition module is used for：
Obtain the eyes image data for first user that the camera of first user takes；
According to the eyes image data of first user, the expression in the eyes direction character data of first user are obtained；
The expression in the eyes direction character data of first user are mapped to the head mould of the virtual portrait of first user in real time
Type obtains the video data of first user.
21. device according to claim 12, which is characterized in that the video data acquisition module is used for：
The behavioural characteristic data of virtual portrait and first user based on first user obtain first user's
Initial video data；
Obtain the perspective data of the second user；
The visual angle indicated by perspective data according to the second user, handles the initial video data, obtain with
The video data of matched first user in visual angle.
22. device according to claim 21, which is characterized in that
The video data acquisition module is used for：According to the collected head orientation characteristic of the sensor of the second user
According to obtaining the corresponding perspective data of head orientation characteristic of the second user；Or,
The video data acquisition module is used for：The eyes image data taken according to the camera of the second user, are obtained
The expression in the eyes direction character data for taking the second user obtain described according to the expression in the eyes direction character data of the second user
The perspective data of two users.
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710104442.4A CN108513089B (en) | 2017-02-24 | 2017-02-24 | Method and device for group video session |
| PCT/CN2018/075749 WO2018153267A1 (en) | 2017-02-24 | 2018-02-08 | Group video session method and network device |
| TW107106428A TWI650675B (en) | 2017-02-24 | 2018-02-26 | Method and system for group video session, terminal, virtual reality device and network device |
| US16/435,733 US10609334B2 (en) | 2017-02-24 | 2019-06-10 | Group video communication method and network device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710104442.4A CN108513089B (en) | 2017-02-24 | 2017-02-24 | Method and device for group video session |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108513089A true CN108513089A (en) | 2018-09-07 |
| CN108513089B CN108513089B (en) | 2021-02-12 |
Family
ID=63373821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710104442.4A Active CN108513089B (en) | 2017-02-24 | 2017-02-24 | Method and device for group video session |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108513089B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109857311A (en) * | 2019-02-14 | 2019-06-07 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, terminal and the storage medium of human face three-dimensional model |
| CN109857249A (en) * | 2019-01-09 | 2019-06-07 | 重庆爱奇艺智能科技有限公司 | It is a kind of for generating the method and apparatus of avatar image |
| CN109886144A (en) * | 2019-01-29 | 2019-06-14 | 深圳市云之梦科技有限公司 | Virtual examination forwarding method, device, computer equipment and storage medium |
| CN110147729A (en) * | 2019-04-16 | 2019-08-20 | 深圳壹账通智能科技有限公司 | User emotion recognition methods, device, computer equipment and storage medium |
| CN110944142A (en) * | 2018-09-21 | 2020-03-31 | 雅马哈株式会社 | Image processing apparatus, camera apparatus, and image processing method |
| CN111953922A (en) * | 2019-05-16 | 2020-11-17 | 南宁富桂精密工业有限公司 | Face identification method for video conference, server and computer readable storage medium |
| CN112435326A (en) * | 2020-11-20 | 2021-03-02 | 深圳市慧鲤科技有限公司 | Printable model file generation method and related product |
| CN112686965A (en) * | 2020-12-25 | 2021-04-20 | 百果园技术(新加坡)有限公司 | Skin color detection method, device, mobile terminal and storage medium |
| CN113538455A (en) * | 2021-06-15 | 2021-10-22 | 聚好看科技股份有限公司 | Three-dimensional hairstyle matching method and electronic equipment |
| CN117727303A (en) * | 2024-02-08 | 2024-03-19 | 翌东寰球(深圳)数字科技有限公司 | Audio and video generation method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102164265A (en) * | 2011-05-23 | 2011-08-24 | 宇龙计算机通信科技(深圳)有限公司 | Method and system of three-dimensional video call |
| CN103238317A (en) * | 2010-05-12 | 2013-08-07 | 布鲁珍视网络有限公司 | Systems and methods for scalable distributed global infrastructure for real-time multimedia communication |
| US20140085406A1 (en) * | 2012-09-27 | 2014-03-27 | Avaya Inc. | Integrated conference floor control |
| CN105721821A (en) * | 2016-04-01 | 2016-06-29 | 宇龙计算机通信科技(深圳)有限公司 | Video calling method and device |
-
2017
- 2017-02-24 CN CN201710104442.4A patent/CN108513089B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103238317A (en) * | 2010-05-12 | 2013-08-07 | 布鲁珍视网络有限公司 | Systems and methods for scalable distributed global infrastructure for real-time multimedia communication |
| CN102164265A (en) * | 2011-05-23 | 2011-08-24 | 宇龙计算机通信科技(深圳)有限公司 | Method and system of three-dimensional video call |
| US20140085406A1 (en) * | 2012-09-27 | 2014-03-27 | Avaya Inc. | Integrated conference floor control |
| CN105721821A (en) * | 2016-04-01 | 2016-06-29 | 宇龙计算机通信科技(深圳)有限公司 | Video calling method and device |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110944142A (en) * | 2018-09-21 | 2020-03-31 | 雅马哈株式会社 | Image processing apparatus, camera apparatus, and image processing method |
| CN109857249A (en) * | 2019-01-09 | 2019-06-07 | 重庆爱奇艺智能科技有限公司 | It is a kind of for generating the method and apparatus of avatar image |
| CN109886144B (en) * | 2019-01-29 | 2021-08-13 | 深圳市云之梦科技有限公司 | Virtual trial sending method and device, computer equipment and storage medium |
| CN109886144A (en) * | 2019-01-29 | 2019-06-14 | 深圳市云之梦科技有限公司 | Virtual examination forwarding method, device, computer equipment and storage medium |
| CN109857311A (en) * | 2019-02-14 | 2019-06-07 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, terminal and the storage medium of human face three-dimensional model |
| CN110147729A (en) * | 2019-04-16 | 2019-08-20 | 深圳壹账通智能科技有限公司 | User emotion recognition methods, device, computer equipment and storage medium |
| CN111953922A (en) * | 2019-05-16 | 2020-11-17 | 南宁富桂精密工业有限公司 | Face identification method for video conference, server and computer readable storage medium |
| CN111953922B (en) * | 2019-05-16 | 2022-05-27 | 南宁富联富桂精密工业有限公司 | Face identification method for video conference, server and computer readable storage medium |
| CN112435326A (en) * | 2020-11-20 | 2021-03-02 | 深圳市慧鲤科技有限公司 | Printable model file generation method and related product |
| CN112686965A (en) * | 2020-12-25 | 2021-04-20 | 百果园技术(新加坡)有限公司 | Skin color detection method, device, mobile terminal and storage medium |
| CN113538455A (en) * | 2021-06-15 | 2021-10-22 | 聚好看科技股份有限公司 | Three-dimensional hairstyle matching method and electronic equipment |
| CN113538455B (en) * | 2021-06-15 | 2023-12-12 | 聚好看科技股份有限公司 | Three-dimensional hairstyle matching method and electronic equipment |
| CN117727303A (en) * | 2024-02-08 | 2024-03-19 | 翌东寰球(深圳)数字科技有限公司 | Audio and video generation method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108513089B (en) | 2021-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108513089A (en) | The method and device of group's video session | |
| US10169905B2 (en) | Systems and methods for animating models from audio data | |
| US10559111B2 (en) | Systems and methods for generating computer ready animation models of a human head from captured data images | |
| CN101324961B (en) | Human face portion three-dimensional picture pasting method in computer virtual world | |
| CN104217454B (en) | A kind of human face animation generation method of video drive | |
| CN107180446B (en) | Method and device for generating expression animation of character face model | |
| US9959453B2 (en) | Methods and systems for three-dimensional rendering of a virtual augmented replica of a product image merged with a model image of a human-body feature | |
| JP3984191B2 (en) | Virtual makeup apparatus and method | |
| CN107274466A (en) | The methods, devices and systems that a kind of real-time double is caught | |
| CN109035373A (en) | The generation of three-dimensional special efficacy program file packet and three-dimensional special efficacy generation method and device | |
| JPWO2008102440A1 (en) | Makeup face image generating apparatus and method | |
| CN104331564B (en) | Adorn guidance method and terminal device based on terminal device | |
| JPWO2018079255A1 (en) | Image processing apparatus, image processing method, and image processing program | |
| US10650564B1 (en) | Method of generating 3D facial model for an avatar and related device | |
| CN107679519A (en) | A kind of multi-modal interaction processing method and system based on visual human | |
| Spencer | Zbrush digital sculpting human Anatomy | |
| EP4160545A1 (en) | Three-dimensional avatar generation device, three-dimensional avatar generation method, and three-dimensional avatar generation program | |
| KR101516465B1 (en) | Method and apparatus for making avatar | |
| JP2024506170A (en) | Methods, electronic devices, and programs for forming personalized 3D head and face models | |
| Maraffi | Maya character creation: modeling and animation controls | |
| CN106446207B (en) | Makeups library banking process, personalized makeups householder method and its device | |
| CN103258065A (en) | System and method for digitalized human body emulation of natural person | |
| WO2020085922A1 (en) | Digital character blending and generation system and method | |
| CN109242982A (en) | A kind of Network Three-dimensional AR dynamic template scene technology | |
| Scheepers | Anatomy-based surface generation for articulated models of human figures |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |